Music may enhance growth and reduce the length of stay in premature infants; early exposure to music may also have long-term effects. However, the pathways by which music affects clinical outcomes remain unknown. The purpose of this study is to evaluate the effects of harp music on convalescing premature infants physiologically through heart rate variability (HRV) and salivary cortisol (SC) and behaviorally through activity level (ACT). We hypothesize that: a) these outcomes will be affected by harp music; b) such effects will be greater for the third dose of music than the first dose; and c) that the three outcomes (HRV, SC and ACT) will be moderately correlated with each other. Subjects will include 90 infants, between 34 and 37 weeks post-conceptional age, hospitalized at our Intermediate Care Nursery (ICN) who have passed the newborn hearing screening test, are medically stable, have been hospitalized and are expected to remain for 10 more days. At baseline, we will measure HRV, SC and activity twice, separated by 45 minutes. Subjects will be randomized to a control group (usual care) OR one of two interventions a) quiet room or b) harp music. Interventions will be given for 45 minutes daily on three consecutive days in the same study room. Primary outcomes include changes in: 1) HRV, 2) SC and 3) ACT. Based on our previous research, the five primary parameters for HRV analysis include the standard deviation of the interbeat interval, sample entropy, sample asymmetry, Hurst parameter, and the HF/LF ratio of the power spectral analysis. Movement will be measured in movements per minute as recorded electronically on an actimeter. Statistical analysis will include: 1) comparisons with baseline of the primary outcomes for usual care control vs. music vs. quiet on Day 1 (effect of first dose of harp music), 2) comparisons of the primary outcomes on Day 1 (first dose) vs. Day 3 (3rd dose), and 3) correlations between the three primary outcome measures at baseline and with the different interventions. Secondary analysis will explore which of the HRV parameters is most sensitive to music in order to determine which is most suitable for future studies. This study provides crucial comparative data for understanding the physiologic and behavioral effects of music vs. decreased noise on premature infants. Future studies will assess the immediate and long-term impact of different kinds, amounts, frequency and duration of music on physiologic, clinical, behavioral and developmental outcomes in premature infants.